分类错误是导致街景分割地理偏差的主因，而非定位问题。
Classification Drives Geographic Bias in Street Scene Segmentation
- 通过分析欧洲数据训练的模型，发现分类错误引发地理偏差。
- 分类错误贡献了分割任务10%-90%、检测任务19%-88%的地理偏差。
- 合并细粒度类别可显著降低分类带来的地理偏差，适合区域专用模型优化。
以往研究显示，缺乏地理多样性的图像数据集会导致模型性能出现偏差。尽管早期工作聚焦于通用图像数据集（如ImageNet）和简单任务（如图像识别），本文则深入探讨了真实驾驶场景数据集中存在的地理偏差，针对更复杂的实例分割任务展开分析。我们检验了在欧洲驾驶场景数据上训练的模型（欧系模型）是否存在地理偏差。结果表明，欧系模型确实存在地理偏差。有趣的是，这些偏差主要源于分类错误，而非定位错误：仅分类错误就占到了分割任务中10%-90%的地理偏差，检测任务中19%-88%的地理偏差。这说明虽然分类任务具有地理敏感性，但定位（包括检测与分割）本身在地理上相对稳健。研究还发现，在特定区域模型中，通过使用更粗粒度的类别（如将汽车、公交车、卡车合并为四轮车）可有效缓解分类错误引发的地理偏差。
原文摘要 · Abstract (English)
Previous studies showed that image datasets lacking geographic diversity can lead to biased performance in models trained on them. While earlier work studied general-purpose image datasets (e.g., ImageNet) and simple tasks like image recognition, we investigated geo-biases in real-world driving datasets on a more complex task: instance segmentation. We examined if instance segmentation models trained on European driving scenes (Eurocentric models) are geo-biased. Consistent with previous work, we found that Eurocentric models were geo-biased. Interestingly, we found that geo-biases came from classification errors rather than localization errors, with classification errors alone contributing 10-90% of the geo-biases in segmentation and 19-88% of the geo-biases in detection. This showed that while classification is geo-biased, localization (including detection and segmentation) is geographically robust. Our findings show that in region-specific models (e.g., Eurocentric models), geo-biases from classification errors can be significantly mitigated by using coarser classes (e.g., grouping car, bus, and truck as 4-wheeler).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。